Lead AI Engineer

Description

The role is a hands-on engineering lead position for agentic AI products, covering implementation patterns and technical quality from prototype through to production.

The team works with clients across sectors, alongside architects, MLOps staff and product owners.

Contribution to the practice is part of the role, including proposals, client workshops and market development at a level suited to your seniority.

Responsibilities

• Lead the technical design and build of agents, retrieval-augmented generation services, tool integrations and model orchestration.
• Break architecture into engineering tasks and help with estimates and sprint planning.
• Set standards for coding, testing, evaluation, code review and documentation.
• Coach engineers, review their code and work through hard technical problems.
• Build evaluation suites that measure quality, safety, reliability, latency and cost.
• Work with architects and MLOps to make solutions ready for production.
• Share reusable methods, patterns, code, assets and lessons learned, and coach colleagues.

Requirements

• At least 6 years in software, data or machine-learning engineering, with hands-on AI delivery.
• Strong Python and API engineering skills, and experience with modern agent or LLM frameworks.
• Experience with retrieval, embeddings, vector stores, model evaluation and distributed systems.
• Ability to lead agile teams while staying hands-on.
• Production experience with Python and API work, using agent frameworks such as LangChain, LangGraph, Microsoft Agent Framework, Semantic Kernel, OpenAI Agents SDK, AutoGen, CrewAI or an equivalent.
• Ability to build graph-based and code-first orchestration covering state, memory, checkpoints, tool calling, hand-offs, retries, idempotency, human approval and long-running workflows.
• Advanced knowledge of RAG, structured outputs, prompt and context engineering, embeddings, vector or hybrid retrieval, reranking, knowledge graphs and retrieval evaluation.
• Experience connecting agents to enterprise systems through REST or GraphQL APIs, events, queues, databases and MCP-compatible tools or servers.
• Practical experience with automated evaluation and observability tools such as LangSmith, MLflow, Langfuse, OpenTelemetry, Azure AI evaluation or equivalent, measuring quality, trajectory, latency, token use and cost.
• Solid software engineering practice: pytest or equivalent testing, type checking, code review, dependency management, secure coding, CI/CD and containerised deployment.

Conditions

Hybrid working.

Your success is judged on delivery pace and predictability, code quality and test coverage, evaluation results and production readiness, fewer defects and less rework, and reusable components built.

You will work with colleagues in the AI Transformation and Agentic Systems Practice, the company specialists in sector, function, cloud, cyber, risk, Responsible AI and change, client business owners and technology teams, and alliance partners where relevant.

You are expected to uphold the company's quality, independence, confidentiality and risk management requirements.